A new AI model developed by a team of researchers in Canada can accurately predict whether a patient will survive cancer by reading notes from their doctor. Photo courtesy of Wikimedia Commons
March 2 (UPI) -- A new AI model developed by a team of researchers in Canada can accurately predict whether a patient will survive cancer by reading notes from their doctor.
Researchers with the University of British Columbia and BC Cancer developed the AI model for part of a study published Thursday in the JAMA Network Open medical journal.
"Our results suggest it is possible to predict the survival of patients with cancer without having to construct structured data sets, or limiting the predictions to specific types or locations of cancer," the authors wrote in the study.
"The availability of structured data may vary. Given the widespread availability of initial oncologist consultation documents, this opens up the possibility of more easily training and using such models across cancer types at different cancer centers."
The authors of the study wrote that accurately predicting whether a patient will survive help improve cancer care.
"For example, it might suggest earlier referral to palliative care resources or consideration of more aggressive therapies upfront," the authors wrote.
In the study, the authors noted that "almost all patients" who receive cancer treatment have an initial consultation with their oncologist in which they reveal "many details relevant to survival, for example, tobacco use or marital status, even if the clinic does not routinely store such data."
In order to conduct the study, the researchers trained the AI model using data from 47,625 patients across all six BC Cancer locations. The researchers noted that they excluded any patients who had more than one cancer diagnosis from the study.
"We calculated survival as the number of months between the selected document being generated, and either the patient's recorded death date or April 6, 2022, when mortality data were last extracted from administrative data," the study reads.
"We then produced binary labels for whether a patient survived 6, 36, or 60 months or not."
Researchers have previously used such documents to predict the onset of at least three non-cancer illnesses, according to the study.
"The AI essentially reads the consultation document similar to how a human would read it," Dr. John-Jose Nunez, a psychiatrist and clinical research fellow who served as the study's lead author, said in a news release.
"These documents have many details like the age of the patient, the type of cancer, underlying health conditions, past substance use, and family histories. The AI brings all of this together to paint a more complete picture of patient outcomes."